feat(notifications): add a failed-check threshold for node.down alerts (#6768)

One lost heartbeat on a lossy link flipped a node offline and sent node.down,
and the next probe sent node.up, so a flaky node flooded Telegram, email and
Discord with alert pairs.

A new nodeDownThreshold setting (default 1, today's behaviour) works like
outboundDownThreshold: node.down fires once the failed-check streak reaches it,
once per outage, and node.up only after a node.down was sent. The announcement
is persisted in nodes.down_notified, so a restart inside the threshold window
neither drops the outage's node.down nor re-announces nodes already reported
offline; nodes offline at upgrade are seeded as announced.
This commit is contained in:
Grigoriy
2026-10-24 05:26:13 +03:00
committed by GitHub
parent c3037fc716
commit 2f62dadd25
34 changed files with 379 additions and 35 deletions
+18 -1
View File
@@ -1475,7 +1475,7 @@ func runSeeders(isUsersEmpty bool) error {
}
if empty && isUsersEmpty {
seeders := []string{"UserPasswordHash", "ClientsTable", "InboundClientsArrayFix", "InboundClientTgIdFix2", "InboundClientSubIdFix", "FreedomFinalRulesReverseFix", "FreedomFinalRulesPrivateEgressBlock", "UppercaseFreedomFinalRulesFix", "InboundRealityFinalmaskTcpStrip", "ApiTokensHash", "LegacyProxySettingsCleanup", "OutboundRemovedKeysFix", "FreedomDomainStrategyFix", "DNSOutboundLegacyKeysFix", "DNSOutboundQTypeZeroFix", "WireguardDomainStrategyFix", "XdnsFinalmaskNamesAddrsFix", "UDPHopClientMaskFix", "HostVerifyPeerCertByNameBoolFix", "WireguardPeersToClients", "MtprotoSecretsToClients", "NodeInboundsAdopted", "ResetIpLimitNoFail2ban"}
seeders := []string{"UserPasswordHash", "ClientsTable", "InboundClientsArrayFix", "InboundClientTgIdFix2", "InboundClientSubIdFix", "FreedomFinalRulesReverseFix", "FreedomFinalRulesPrivateEgressBlock", "UppercaseFreedomFinalRulesFix", "InboundRealityFinalmaskTcpStrip", "ApiTokensHash", "LegacyProxySettingsCleanup", "OutboundRemovedKeysFix", "FreedomDomainStrategyFix", "DNSOutboundLegacyKeysFix", "DNSOutboundQTypeZeroFix", "WireguardDomainStrategyFix", "XdnsFinalmaskNamesAddrsFix", "UDPHopClientMaskFix", "HostVerifyPeerCertByNameBoolFix", "WireguardPeersToClients", "MtprotoSecretsToClients", "NodeInboundsAdopted", "NodeDownNotified", "ResetIpLimitNoFail2ban"}
for _, name := range seeders {
if err := db.Create(&model.HistoryOfSeeders{SeederName: name}).Error; err != nil {
return err
@@ -1640,6 +1640,12 @@ func runSeeders(isUsersEmpty bool) error {
}
}
if !slices.Contains(seedersHistory, "NodeDownNotified") {
if err := seedNodeDownNotified(); err != nil {
return err
}
}
if err := seedHostsFromExternalProxy(); err != nil {
return err
}
@@ -1689,6 +1695,17 @@ func seedNodeInboundsAdopted() error {
return db.Create(&model.HistoryOfSeeders{SeederName: "NodeInboundsAdopted"}).Error
}
// seedNodeDownNotified treats nodes already offline at upgrade as announced,
// so the first heartbeat after it does not re-send their node.down.
func seedNodeDownNotified() error {
if err := db.Model(&model.Node{}).
Where("status = ?", "offline").
Update("down_notified", true).Error; err != nil {
return err
}
return db.Create(&model.HistoryOfSeeders{SeederName: "NodeDownNotified"}).Error
}
// backfillEmptyHostGroupIds is idempotent and not seeder-gated: builds that
// predate group ids on the inbound-import path (and restored backups) can
// re-introduce hosts rows with an empty group_id, and such rows render as a